About the role
You'll be the technical owner of GenAI and agentic AI for a security SaaS product. You'll build the LLM-backed features customers use in production, not internal tools or demos. As the first and only AI voice on the team, you'll make the architecture calls yourself.
Responsibilities:
- Design, build and ship RAG pipelines and multi-agent systems in production.
- Build Python/FastAPI backend services that serve LLM features to enterprise customers.
- Build and integrate MCP servers and tool-calling workflows.
- Set up evaluation and observability for LLM outputs, with eval running in CI/CD.
- Work directly with the founders and take ownership with minimal spec.
Requirements:
- Python with FastAPI or Flask as your primary backend stack.
- Hands-on experience with LangChain, LangGraph or a similar agent framework in real project work.
- A RAG pipeline you personally built, with a vector DB (Pinecone, Qdrant, Milvus, FAISS, PGVector or ChromaDB).
- 10+ years of total engineering experience.
- 3+ years of hands-on GenAI/LLM work in production.
- Experience at an enterprise SaaS product company (not services or consulting).
- Based in Bengaluru, or willing to relocate for 5 days a week in the office.
- Actively looking for a change.
Good to have:
- MCP built or integrated in production.
- RAGAS, DeepEval or LangSmith for evaluation.
- Security, identity or compliance domain exposure.
- Early-stage startup or 0-to-1 build experience.
- AWS Bedrock or Google Vertex AI.
- Multi-agent frameworks (AutoGen, CrewAI, Semantic Kernel, ADK).
Requirements
- Python with FastAPI or Flask as your primary backend stack.
- Hands-on experience with LangChain, LangGraph or a similar agent framework in real project work.
- A RAG pipeline you personally built, with a vector DB (Pinecone, Qdrant, Milvus, FAISS, PGVector or ChromaDB).
- 10+ years of total engineering experience.
- 3+ years of hands-on GenAI/LLM work in production.